tahrirchi/uz-books

Dataset Card for BookCorpus

Dataset Summary

In an effort to democratize research on low-resource languages, we release UzBooks dataset, a cleaned book corpus consisting of nearly 40000 books in Uzbek Language divided into two branches: "original" and "lat," representing the OCRed (Latin and Cyrillic) and fully Latin versions of the texts, respectively.

Please refer to our blogpost and paper (Coming soon!) for further details.

To load and use dataset, run this script:

from datasets import load_dataset

uz_books=load_dataset("tahrirchi/uz-books")

Dataset Structure

Data Instances

plain_text

• Size of downloaded dataset files: 16.98 GB
• Size of the generated dataset: 32.95 GB
• Total amount of disk used: 49.93 GB

An example of 'train' looks as follows.

{
    "text": "Hamsa\nAlisher Navoiy ..."
}

Data Fields

The data fields are the same among all splits.

plain_text

• text: a string feature that contains text of the books.

Data Splits

name
original	39712
lat	39712

Dataset Creation

The books have been crawled from various internet sources and preprocessed using Optical Character Recognition techniques in Tesseract OCR Engine. The latin version is created by converting the original dataset with highly curated scripts in order to put more emphasis on the research and development of the field.

Citation

Please cite this model using the following format:

@online{Mamasaidov2023UzBooks,
    author    = {Mukhammadsaid Mamasaidov and Abror Shopulatov},
    title     = {UzBooks dataset},
    year      = {2023},
    url       = {https://huggingface.co/datasets/tahrirchi/uz-books},
    note      = {Accessed: 2023-10-28}, % change this date
    urldate   = {2023-10-28} % change this date
}

Gratitude

We are thankful to these awesome organizations and people for helping to make it happen:

• Ilya Gusev: for advise throughout the process
• David Dale: for advise throughout the process

Contacts

We believe that this work will enable and inspire all enthusiasts around the world to open the hidden beauty of low-resource languages, in particular Uzbek.

For further development and issues about the dataset, please use [email protected] or [email protected] to contact.